Volume 43, Issue 5, Pages 987-997 (November 2015) Inflammasomes Coordinate Pyroptosis and Natural Killer Cell Cytotoxicity to Clear Infection by a Ubiquitous Environmental Bacterium  Vivien I. Maltez, Alan L. Tubbs, Kevin D. Cook, Youssef Aachoui, E. Liana Falcone, Steven M. Holland, Jason K. Whitmire, Edward A. Miao  Immunity  Volume 43, Issue 5, Pages 987-997 (November 2015) DOI: 10.1016/j.immuni.2015.10.010 Copyright © 2015 Elsevier Inc. Terms and Conditions

Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 1 Inflammasomes and the NADPH Oxidase Are Required to Clear Specific CGD Pathogens (A–D) Mice were infected i.p. with (A) 2 × 107 or (B) 1 × 104 CFU of B. thailandensis, (C) 1 × 106 CFU of B. cepcacia, (D) or 1 × 106 CFU of F. philomiragia and monitored for survival. Casp1–/–Casp11–/– mice referred to as Casp1-11DKO. Mice: (A) WT = 9, Casp1-11DKO = 10, Ncf1–/– = 10; (B) WT = 4, Casp1-11DKO = 6, Ncf1–/– = 9; (C) n = 4 for all; (D) n = 5 for all. (E) Mice were infected i.p. with C. violaceum (Cv) at the indicated doses and monitored for survival. Mice: WT = 4, Casp1-11DKO = 7, Ncf1–/– = 4. (F–I) Mice were infected i.p. with 100 CFUs of Cv and bacterial burdens were measured in the liver (F and H) and spleen (G and I) 72 hr later (F and G) or 17 hr later (H and I). (J) Mice were infected i.p. with 100 CFUs of Cv and 72 hr later livers and spleens were embedded in paraffin and H&E stained. Data and statistical analyses are representative of at least two experiments (A–D, H–I), three experiments (F–G), or pooled from two experiments (E). Dashed line indicates limit of detection. ∗p ≤ 0.05; survival curve analysis (A–E), two-tailed unpaired t test (F–I). See also Figure S1. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 2 Canonical Inflammasome Signaling via NLRC4 Prevents Lethal Infection by C. violaceum In Vivo (A and B) Mice were infected i.p. with 100 CFUs of Cv and monitored for survival. Nlrc4–/–Asc–/– mice referred to as Nlrc4-AscDKO. Mice: (A) Casp1-11DKO = 4, Nlrc4-AscDKO = 4, Casp11–/– = 5; (B) n = 4 for all. (C–F) Mice were infected i.p. with 100 CFUs of Cv, and bacterial burdens were measured in livers (C and E) and spleens (D and F) 72 hr later. All data and statistical analyses are representative of at least three experiments. ∗p ≤ 0.05; survival curve analysis (A and B); two-tailed unpaired t test (C–F). See also Figure S2. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 3 Differential Role for Pyroptosis in the Liver and Spleen (A and B) Mice were infected with 100 CFUs (A) or 104 CFUs (B) Cv and monitored for survival. Il1b–/–Il18–/– mice referred to as Il1b-Il18DKO. Mice: (A) Casp1-11DKO = 5, Il1b-Il18DKO = 7, Il1b–/– = 5; (B) Il1b-Il18DKO = 5, Il1b–/– = 5, Il18–/– = 4, Asc–/– = 5. (C–F) Mice were infected i.p. with 100 CFUs of Cv and bacterial burdens were measured in livers (C and E) and spleens (D and F) 72 hr later. Data and statistical analyses are representative of at least two experiments (A and D–F) or three experiments (B and C) or pooled from two experiments (E and F). ∗p ≤ 0.05; survival curve analysis (A and D), two-tailed unpaired t test (B, C, E, F). Asterisk over a line indicates that all datasets under the line have p ≤ 0.05 for each pair-wise comparison possible. See also Figure S3. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 4 Cytotoxic Activity of NK Cells Is Required for Clearance of C. violaceum (A–D, K, and L) Mice were infected i.p. with 104 CFUs (A, B, K, and L) or 100 CFUs (C and D) Cv at the indicated doses and bacterial burdens were measured in livers (A, C, and K) and spleens (B, D, and L) 72 hr later. (A and B) Mice were injected i.p. with 75 μg of isotype or anti-NK1.1 antibody at 1 and 3 days prior to infection. (E) Primary hepatocytes and splenocytes from naive WT mice were treated with LPS, lysed, and examined for caspase-1 via immunoblot. (F) Mice were infected i.p. with 100 CFUs of Cv and bacterial burdens were measured in livers 72 hr later. Equal weight liver sections were removed and the hepatocyte fraction was graphed as a percentage of the total burden by weight. (G) Primary hepatocytes were infected with Cv at an MOI of 25 and percentage of invasive bacteria was determined via gentamycin protection assay. (H) Primary hepatocytes, liver immune cells, and splenocytes were infected with Cv at an MOI of 10 and treated with gentamycin, and LDH release was determined (note, different splenocytes have a distinct LDH content per cell, thus in this case LDH release does not directly translate to percent lysis). (I and J) Mice were infected i.p. with 103 CFUs of L. monocytogenes (Lm) and bacterial burdens were measured in livers (I) and spleens (J) 72 hr later. Data and statistical analyses are representative of two experiments (C, D, I, J) or pooled from two (K, L) or three (A, B) experiments. ∗p ≤ 0.05; two-tailed unpaired t test (A–D and I–L). Error bars in (F)–(H) indicate SEM. See also Figure S4. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 5 Adaptive Immune Responses Are Not Required for Sterilizing Immunity against C. violaceum Mice were infected i.p. with 103 CFUs of Cv and bacterial burdens were measured in livers (A) and spleens (B) at the indicated times. Data and statistical analyses are representative of two experiments. ∗p ≤ 0.05; two-tailed unpaired t test. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 6 Exogenous IL-18 Recruits NK Cells and Primes Their Cytotoxic Effects (A–G) Mice were infected i.p. with 100 CFUs of Cv and bacterial burdens were measured in livers (A, F, and G) and spleens (B) 72 hr later. (A, B, and F) Mice were injected i.p. with PBS or recombinant mouse IL-18 (0.2 μg/dose at 0, 24, and 48 hr). (F) Mice were injected i.p. with 75 μg of isotype or anti-NK1.1 antibody at 1 and 3 days prior to infection. (C–E) Liver immune cells were collected and prepared for flow cytometry 1 day after infection (C). Markers for conventional NK (NK1.1hiDX5+), liver-resident NK (NK1.1hiDX5–), and (probably) NKT (NK1.1+DX5–CD3+) cell populations (D) were determined in addition to granzyme B staining (E). (H) Hepatocytes were harvested from Ncf1–/– mice infected with 100 Cv 17 hpi. Liver immune cells were harvested from WT mice infected with 104 Cv under three conditions: isotype antibody treatment, anti-NK1.1 antibody treatment, or IL-18 therapy. Cells were coincubated in a 96-well plate and hepatocyte killing was determined by LDH assay 6 hr later. A model of the experimental setup is provided. Data and statistical analyses are representative of at least two experiments (A–F) or pooled from two experiments (F). ∗p ≤ 0.05; two-tailed unpaired t test (A, B, and D–H). Error bars in (D), (E), and (H) indicate SEM. See also Figure S5. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 7 L. monocytogenes Burdens Are Reduced upon IL-18 Administration Mice were infected i.p. with 103 CFUs of Lm and injected i.p. with PBS or recombinant mouse IL-18 (24, 48, and 72 hr), and bacterial burdens were measured in WT and Prf1−/− mice (A and B), Ifng−/− mice (C and D), and Rag1−/− mice (E and F) livers (A, C, and E) and spleens (B, D, and F) 4 days later. Data and statistical analyses are pooled from two experiments. ∗p ≤ 0.05; two-tailed unpaired t test. Immunity 2015 43, 987-997DOI: (10.1016/j.immuni.2015.10.010) Copyright © 2015 Elsevier Inc. Terms and Conditions